One of the great promises of the CenSoc datasets is the opportunity to pull back the curtain a bit more on how Americans in the twentieth century lived before they died. On behalf of the CenSoc team, I am excited to announce what may be the first empirical paper to use CenSoc data in that vein. Published in Population and Environment earlier this month, the paper is entitled, “Does a prolonged hardship reduce life span? Examining the longevity of young men who lived through the 1930s Great Plains drought.”

The title may be a mouthful— such is the agony in trying to lure a reader with a title that may become their only exposure to your work—but it describes a question well suited to CenSoc data. (In fact, the analysis contained in the paper wouldn’t have been possible before CenSoc.) Can something that stresses a group of people for years wear those people down such that their biological aging is permanently accelerated? A lot of extant work demonstrates the link between early childhood conditions and later-life outcomes, including mortality. Yet adverse exposures can occur at any time, and it’s reasonable to expect chronically negative experiences in early adulthood to have long-lasting effects.

For young men residing in the Southern Plains throughout the 1930s, the paper shows that exposure to drought, dust storms, and economic ruin did not reduce their average life span relative to young men in the less-droughted Ohio Valley—in fact, it was just the opposite; the average age at death was over half a year greater for drought-exposed Plainsmen. Moreover, in comparing the life spans of Plainsmen who stayed in the droughted, windblown region for the entire decade and those who left partway through, results suggest the stayers may have been slightly longer-lived than the leavers.

These findings run counter to the expectations laid out in the first part of the article, but rejecting a well-reasoned hypothesis is never a bad thing if it leads to other equally important questions. In this paper, I suggest that the longevity advantage among drought survivors may reflect the winnowing of weaker Plainsmen dying before age 65 (the lower bound of the study’s observed ages at death) or the unseen salutary effects of drought recovery programs on stayers. Either one represents an important avenue for future research.

Ultimately, CenSoc’s linking of the 1940 Census to millions of Social Security death records gives population scientists another means to examine how the past influenced the future. If the 1940 Census was the first detailed snapshot of the American population’s home and work life, and death is the ultimate individual-level absorbing state, the CenSoc datasets ask us to ponder, how did we get from there to here? Just as importantly, they allow us to investigate why the ultimate outcome (death) came sooner for some people than for others. What disparities might have caused or underlain this variability?

This paper may have been the first empirical analysis out of the CenSoc gate, but it will be far from the last. You can access it for free here.

Story by Serge Atherwood (satherwood@berkeley.edu).